The present invention relates to the construction of libraries of polymers based on the design principle of introducing systematic variations in the structure of a copolymer in at least two separate and independently variable domains within the copolymer structure. The inventive methodology is broadly applicable to the development of copolymers where complex requirements necessitate a careful optimization of copolymer structure.
This invention is also related to the preparation of tyrosine-derived monomers as disclosed in U.S. Pat. No. 5,587,507 and to the preparation and use of polyarylates as disclosed in U.S. Pat. Nos. 5,216,115 and 5,317,077, which are hereby incorporated by reference.
Combinatorial approaches that have led to dramatic changes in the way lead compounds for the discovery of new drugs are identified are disclosed by Lowe, JCS Reviews, 309-317 (1995). As usually practiced in the pharmaceutical industry, combinatorial schemes are used to create a large number of structurally related compounds (almost always within a single reaction vessel) followed by the identification of potential lead compounds in a selective bioassay. Such combinational schemes are disclosed by Mischer, ChemTracts-Org. Chem., 8(1) 19-25 (1995). This approach is not readily applicable to the design of engineering polymers or biomaterials. Starting with a mixture of monomers and creating a large number of different polymers within the same reaction vessel would result in a blend of polymers that would be difficult to resolve into individual compounds. While such approaches are disclosed by EP 789,671 to be useful in the design of polymers with catalytic activities, these approaches are not useful for the design of polymers where individual material properties need to be optimized.
Correlations between chemical structure and the properties of polymers were explored since the early 1930's when the macromolecular structure of polymers was first recognized. Sometimes, materials are studied in “sets” of structurally unrelated materials. For example, studying the tensile strength of glass, iron, paper, wood, and polyethylene may lead to the identification of a material with suitable strength for any given application, however, because of the lack of any systematic variation in structure between the test materials, it is impossible to draw any generally useful conclusions from such a study.
A study design of somewhat greater sophistication is illustrated by Ertel et al., J. Biomed. Mater. Res., 28, 919-930 (1994). A group of four polymers was investigated that had identical backbone structures but differed in the length of an alkyl ester pendent chain attached to the polymer backbone at each repeat unit. In this study, a relatively small number of polymers were compared and valid conclusions were drawn regarding the effect of increasing pendent chain length on selected polymer properties such as glass transition temperature, rate of chemical degradation etc. It is estimated that hundreds of studies of this kind have been reported in the literature. The main limitation of this study design is that only one single structural variable can be explored.
A study design of greater complexity attempts to correlate the effect of two or more structural variables on a set of selected materials properties. For example in the field of poly(acrylic acid) derivatives, a limited number of studies attempted to correlate the effect of simultaneous variations within the chemical structure of the acrylate pendent chains. Although such studies are known in the literature, the general paradigm of combinatorial synthesis has never been applied to the combinatorial synthesis of copolymers with systematic and defined structural features. In terms of the requirements for library design, this approach is unknown in the prior art.
Menger et al., J. Org. Chem., 60, 6666-6667 (1995), exposed a preformed polymer with reactive pendent chains to a reaction mixture that contained a series of different reactants. As a random coupling procedure was initiated, random sequences of pendent chains were attached to the preexisting polymer backbone. In essence, this is an approach whereby an untraceable mixture of pendent chain sequences were all prepared at once. The individual sequences could not be isolated and no structurally defined materials were obtained. Rather, the entire mixture was tested for a specific catalytic activity and it was impossible to detect which particular pendent chain sequence was responsible for any observed catalytic activity.
Considering the cost and time required to identify carefully designed and optimized polymeric materials as “specialty polymers” in many different industrial, medical, and scientific applications, there is a great need to develop new paradigms and approaches that can (1) increase the number of candidate polymers available for any specific application and (2) systematize the study of correlations between polymer structure on one hand and material properties and performance on the other hand.